gallivm: use f16c hw support for float->half and half->float conversion
authorRoland Scheidegger <sroland@vmware.com>
Tue, 2 Apr 2013 23:06:52 +0000 (01:06 +0200)
committerRoland Scheidegger <sroland@vmware.com>
Wed, 3 Apr 2013 23:03:42 +0000 (01:03 +0200)
commit067a0ae42017f327edce7634890e699b4ec8492c
tree0b5306641ea0aec8efe71c58f87a456de295d756
parent302df7cc85b0e2ce47c40048f30bd116b0d692fc
gallivm: use f16c hw support for float->half and half->float conversion

Should be way faster of course on cpus supporting this (includes AMD
Bulldozer and Jaguar cores, Intel Ivy Bridge and up (except budget models)).
Passes piglit fbo-blending-formats GL_ARB_texture_float -auto on Ivy Bridge.

Reviewed-by: Brian Paul <brianp@vmware.com>
src/gallium/auxiliary/gallivm/lp_bld_conv.c
src/gallium/auxiliary/gallivm/lp_bld_init.c
src/gallium/auxiliary/util/u_cpu_detect.c
src/gallium/auxiliary/util/u_cpu_detect.h